Abstract
An experimental campaign was carried out to study the passivity of steel embedded in Ordinary Portland mortar. Mockup samples made of steel coupons (roughly polished or pre-corroded) embedded in mortar were tested under a relative humidity of 80, 90 and 95% or in a solution simulating the ground water of the French site of Bure (at 25 and 50°C) and under aerated and de-aerated conditions, during up to 3 years. The results of gravimetric measurements show that the average corrosion rates determined on polished coupons are typical of passive conditions (<1 μm/year) and vary from 0·2 to 2·3 μm/year for pre-corroded coupons. Moreover, characterisation (Raman spectroscopy and SEM-EDS) of the samples tested in aerated conditions points out a stable state in term of the nature of the corrosion products and of the composition of the steel/mortar interface. In de-aerated conditions, magnetite is observed in the corrosion products.
Introduction
In the framework of the current French disposal concept for medium level radioactive waste, containers and engineered barriers are to be built with reinforced concrete. 1 Owing to the long term implications of these applications (up to several hundred years), a specific interest in passive corrosion arises. In the high alkalinity of the interstitial solution of the concrete, the passive corrosion is promoted by the formation of a thin protective layer. The stability of this layer depends highly on the solution chemistry (pH, buffering effect, carbonate and sulfate content). 2 The passive film is often described as a double layer composed of a continuous inner layer and a outer layer more loosely bound.3,4 Modeling of mild steel passive corrosion is based on the growth of the dense inner layer considered to be limited by the motion of species (electron or point defect motion) in the solid phase.5,6,7 Electrochemical impedance spectroscopy analysis of the anodic growth of passive layers show results in accordance with the point defect model theory. 8 The composition of the passive film was experimentally characterised.9–14 Several authors reported the presence of Fe2+ in the inner layer.9–11 The passive film has also been described as based on a magnetite type structure 13 or composed of oxyhydroxyde and maghemite. 14
The study of steel embedded in cementitious materials, and not just immersed in representative solutions, is a rather original approach. 15 The present paper sums up the analytical results (gravimetric measurements, Raman spectroscopy and SEM-EDS) obtained on steel coupons embedded in ordinary Portland mortar and aged in various environmental conditions (clay rock environment, humid atmosphere and aerated/de-aerated conditions) during up to three years.
Material and methods
Material
The mockups consist in a steel coupon (16 mm diameter and 5 mm height) embedded in the center of a mortar cylinder (50 mm diameter and 39 mm height). The coupons are made from mild steel rebar and are roughly polished or chemically pre-corroded (immersion in H2SO4–Na2SO4 solution during 24 h). The mortar is prepared by mixing a Val d'Azergues CEM I type cement from Lafarge Company with distilled water (w/c = 0·5) and normalized sand (s/c = 3). The samples are cured during 28 days in their mould at 21°C.
Experimental conditions
Mockups containing polished or pre-corroded steel coupons were tested under nine different environmental conditions. The references associated to these different conditions, and used in this document, are defined in Table 1. Saturated salt solutions (KBr, KNO3 and K2SO4) were used to control the relative humidity (respectively 80, 90 and 95%).The composition of the solution simulating the clay underground water is defined in Table 2 for 25 and 50°C. 16 Experiments in de-aerated environment were carried out in glove boxes, where the oxygen concentration is lower than 3 ppm. One exception is the experiment in Bure site water at 50°C which was performed in a specific device which allows carrying experiments in representative conditions of deep geological storage. This device is composed of four stainless steel cells containing the mockups. At the beginning of the experiment, these cells are filled with synthetic Bure water at 50°C. The temperature is regulated through immersion of the cells in an oil bath at 50°C. The design of this device is similar to the one described in Refs. 17 and 18.
Tested conditions in term of aerated (AE) or de-aerated (DE) environment, temperature, relative humidity (RH) and salt used (to fix RH) or synthetic Bure site water (W): number at end of reference corresponds to relative humidity or to water's temperature
Calculated chemical composition of Bure site water used for experiments on steel coupons embedded in mortar
Analytical methods
Characterisations were performed after 1, 2 and 3 years of simulated aging. Mockups dedicated to surface analysis (Raman spectroscopy and SEM-EDS) were cut and coated in epoxy resin in order to prepare cross-sections showing the steel/mortar interface. Mockups containing polished steel and aged one year are characterised only by XPS. 9
Gravimetric measurements
Corrosion products were removed from the steel surface by chemical attack. The solution used is a mix of hydrochloric acid, antimony trioxide and tin (II) chloride. 19 The difference between the initial mass (weight before pre-corrosion) and the final one (weight after experiment and cleaning) gives the total steel mass loss. In the case of pre-corroded steel coupons, this mass loss must be corrected to take into account the mass loss associated to the corrosion products produced during the pre-corrosion treatment. The latter is considered as 3·8 mg with a standard deviation of 1·9 (value deduced from gravimetric measurements performed on 13 pre-corroded samples). Based on these measurements, the average corrosion rate is estimated considering a uniform corrosion phenomenon.
Micro-Raman spectroscopy
The structural identification of the corrosion products was performed by mean of micro-Raman spectroscopy on the polished cross-sections at the MONARIS lab (UMR 8233). The spectrometer is a Horiba-Jobin Yvon LabRam Infinity and the laser used is a frequency doubled Nd YAG emitting at 532 nm. The setup achieves a spectral resolution of about 2 cm−1. Microanalyses were performed with a long working distance 100x Olympus objective giving an analysed area of about 1·5 μm in diameter. The laser beam is carefully filtered below 100 μW in order to prevent potential thermal transformation of sensitive iron phases. 20 Spectra identification is achieved by comparison with reference spectra given by earlier publications.21–23
Scanning electron microscopy and energy dispersive spectroscopy
Chemical compositions of corrosion products and binder were characterized with two different scanning electron microscopes. The first one is a LEO 120, from Cambridge Instruments, associated with SAMx Max View software (version 4·1·3). The accelerating voltage used is 15 kV. Energy dispersive X-ray spectroscopy analyses were performed using the SAMx IDFix software. The second one is a field emission gun equipped with a hot cathode (MEB-FEG 7000F, JEOL). The acquisition system is a Bruker SDD detector (silicon dry diode) associated with ESPRIT software. Oxygen quantification is made by the use of Si(Li) detectors equipped with a thin beryllium window (∼2 wt-% relative error on iron oxides standards). The lower detection limit is of 0·5 wt-% for elements heavier than aluminum and several percents for oxygen.
Results
Average corrosion rates
Average corrosion rates were determined from the gravimetric measurement performed on polished and pre-corroded steel coupon embedded in mortar and tested under different conditions. For the polished steel coupons, the average corrosion rates obtained are lower than 0·16 μm/year whatever environmental condition and duration (Fig. 1a and b).
a, c results from mockups tested under aerated conditions and relative humidity of 80% (AE-RH 80), 90% (AE-RH 90) and 95% (AE-RH 95) or in Bure site water at 25°C (AE-W 25) and 50°C (AE-W 50); b, d results from mockups tested under de-aerated conditions and relative humidity of 80% (DE-RH 80) and 95% (DE-RH 95) or in Bure site water at 25°C (DE-W 25) and 50°C (DE-W 50)
For pre-corroded steel coupons, the analysis of the gravimetric is trickier because of the pre-corrosion treatment. For mockups tested under humid atmospheres, mass loss measurements show no significant evolution with time, and the mass losses measured (3-7 mg) are close to the one associated to pre-corrosion treatment. After three years of experiment, the average corrosion rates are lower than 1 μm/year (Fig. 1c and d). For experiments in Bure site water, the evolution of mass loss can show discrepancies: the most important is a mass loss 10 mg higher after 1 year than after 2 years in aerated conditions at 25°C. The uncertainties associated to the pre-corrosion treatment could explain that observation. It should also be noted that gravimetric measurements are performed on a unique sample for each condition and ageing duration. Nevertheless, the mass loss results associated to mockups immersed in Bure site water are significantly higher than those obtained in humid atmospheres. The average corrosion rates decrease with time, up to about 1-2 μm/year after 3 years, which could suggest a transient behavior leading towards the passive corrosion state. Besides, the presence of 41 mmol L−1 chloride in Bure water does not seem to induce a significant increase in the corrosion rate. The lack of influence of the thick corrosion products layer (some tens of micrometers in thickness) could suggest that the kinetics is controlled by the thin inner passive layer and not by the dissolution of the outer corrosion products, as proposed by Bataillon et al. 7 Thus, the electrochemical processes within the passive film would be the main ones to be taken into account in the modelling
Whatever the environmental conditions, the corrosion rates determined in this study are of the same order of magnitude that the ones found in literature. In aerated condition, several workers have already estimated the average corrosion rate to be lower than 1 μm/year.24–26 For archeological analogs tested in deuterated iron hydrocarbonate solution under anoxic conditions the corrosion rate is lower than 2 μm/year.27,28
Nature of corrosion products
The results of Raman analysis performed on mockups containing a pre-corroded steel coupon are summarised in Table 3. In aerated environment, for all experimental conditions, the pre-corroded coupons present the same kind of corrosion products than the ones formed during the pre-corrosion treatment (Fig. 2a), namely goethite, ferrihydrite and lepidocrocite. In de-aerated conditions, corrosion products typical of aerated condition are also detected, even though these phases are not stable in anoxic conditions. 29 The procedure of preparation of the mockups and their curing in aerated conditions could explain this observation. Along with the initial corrosion products, the formation of maghemite is suggested by the presence of a double peak at 670 and 710 cm−1. Besides, magnetite seems to appear within these layers when the aging time increases (Fig. 2b). The presence of magnetite in corrosion products has also been observed in a previous study of corrosion in soils. 28 Moreover, the corrosion reaction supplies the elements needed for the reduction of phases containing Fe3+ into magnetite.
Images and Raman spectra of corrosion product on pre-corroded steel coupon embedded in mortar and a immersed in Synthetic Bure site water at 50°C in aerated conditions during 3 years or b tested in de-aerated conditions and relative humidity of 95% for duration of 3 years
Iron oxide and oxihydroxide phases detected by Raman spectroscopy within corrosion product layers formed on pre-corroded steel coupons embedded in mortar tested under different environmental conditions
Steel/mortar interface
SEM-EDS analyses of the steel/mortar interface highlighted two different kinds of corrosion patterns. The first typical corrosion pattern is observed on pre-corroded coupons. The steel/mortar interface is characterised by a succession of two layers: a first layer of corrosion products at the surface of the coupon followed by an iron enriched layer, and finally the unaltered mortar (Fig. 3a). Generally, the corrosion products are contaminated by the major elements of the mortar, i.e. calcium and silicon. The thickness of the layer of corrosion products varies from some micrometers to some tens of micrometers. These values are in the same range than those observed for the corrosion products formed after the pre-corrosion treatment. The second typical corrosion pattern is found in mockups containing polished steel coupon (Fig. 3b) and sometimes in mockups containing pre-corroded steel coupon (areas without corrosion products). This pattern is characterised by an iron enriched layer (named IEL in Fig. 3). In the two different kinds of corrosion patterns, an iron enriched layer is detected. This iron enrichment is associated to a relative depletion of the major elements of mortar, i.e. calcium and silicon. Whatever the environmental conditions and durations, the thickness of the iron enriched layer does not increase significantly between 1 and 3 years of aging. After 3 years, it remains lower than 100 μm (up to 60 μm). It must be noted that the limit between the iron enriched layer and the unaltered mortar can be difficult to determine due to the presence of iron in the mortar.
Elementary composition (wt-%), in function of distance from surface of steel coupon, of principal elements: oxygen (O), silicon (Si), calcium (Ca) and iron (Fe): SEM-EDS analyses are performed a on sample containing pre-corroded steel coupon tested under de-aerated condition and relative humidity of 80% for aging time of 1 year as well as b on polished steel coupon embedded in mortar and tested under aerated condition and relative humidity of 95% for aging time of 2 years; CP and IEL stand for respectively corrosion product and iron enriched layer; CP layer is formed during pre-corrosion treatment and thus can only be observed on pre-corroded steel coupon
These two kinds of configuration were also detected after very short aging times (1 month) on samples containing polished or pre-corroded steel coupons as well as in ‘old corrosion product layer’/mortar interface in aerated environment. 15 After one month of experiment at a relative humidity of 95% in aerated condition, the thickness observed is of the same order of magnitude than the one estimated after a few years (Fig. 4).
Mass content (wt-%) of principal elements, oxygen (O), silicon (Si), calcium (Ca) and iron (Fe), in function of distance from surface of steel coupon; SEM-EDS analyses were performed on mockups containing a pre-corroded steel coupon and b polished steel coupon tested under aerated condition and relative humidity of 95% for aging time of 1 month; CP and IEL stand for respectively corrosion products and iron enriched layer
Conclusions
Corrosion studies on steel coupons embedded in Ordinary Portland mortar and aged under different environmental conditions [under a relative humidity of 80, 90 and 95% or immersed in a solution simulating the clay rock ground water of the French site of Bure (at 25 and 50°C), in both cases under aerated and de-aerated conditions, and for a duration of 1 to 3 years] point out that as follows.
Whatever the environmental conditions, the estimated average corrosion rates deduced from polished steel coupons are typical of a passive corrosion state. For pre-corroded steel coupons, comparable conclusions can be drawn: the passive corrosion state seem to be reached sooner or later whatever the environmental conditions.
The nature of corrosion products present at the pre-corroded steel/mortar interface doesn't seem to evolve with time, except in de-aerated environment where magnetite seems to be detected when the aging time increases.
The steel coupon/mortar interface shows an iron enrichment. The thickness of the iron enriched layer doesn't seem to evolve significantly from curing time to 3 year aging.
All these observations tend to indicate that the corrosion pattern would be stable from 1 month to 3 years for all the tested environmental conditions. Controlled laboratory experiments of longer duration would be necessary to more accurately determine the passive corrosion rate and to conclude on the formation of magnetite in de-aerated conditions.
Footnotes
Acknowledgements
The authors acknowledge the financial support provided by CEA, ANDRA and EDF and Cécile Blanc (CEA - DEN/DPC/SEARS/LISL) for SEM-EDS analyses.